Dr. A. C. Upton (Oak Ridge National Laboratory): Dr. Kaplan has presented an extremely interesting and provocative observation. I was tempted, on reading the abstract, to view the priming effect, so to speak, as an example of tumor initiation and the effects of subsequent irradiation without shielding as promotion.

In an effort to ascertain whether in our system, the RF mouse, radiation might act as a promoting agent as well as an initiating agent, we have been comparing the susceptibility of irradiated adults and nonirradiated adults to the leukemogenic action of filtrates from lymphomatous tissue and tissue of animals with myeloid leukemia. We are getting some indications that the irradiated adult does show an enhanced yield of granulocytic leukemia and lymphoma when injected with filtrates from leukemic donors, which suggests that radiation does change the responsiveness of the host. [Carcinogenesis, Mechanisms of Action. In Ciba Foundation Symposium. (Wolstenholme, G. E. W., and O'Connor, C. M., eds.). London, J. & A. Churchill, Ltd., 1959, pp. 249-268.]

It is distressing to us to find inconsistencies in our results from one experiment to the next. I am wondering what this may denote in terms of activity of passage materials or host sensitivity. Dr. Donald F. Parsons and I have been attempting to screen our passage materials with the electron microscope, hoping possibly to find a correlation between particle content and subsequent leukemogenic activity on inoculation.

In this connection, it is of interest to inquire whether there is any information as yet about the mechanism of the gradual enhancement of tumor-inducing potency of passage materials reported by Dr. Moloney and by Dr. Gross. Are these merely quantitative changes, i.e., increase in titer, or qualitative changes in the agent, i.e., selection or possibly mutation toward increased tumor-producing virulence? These are old questions in virology; I am wondering if we have relevant information on leukemia viruses.

Finally, to return to host resistance, our use of the irradiated adult RF recipient was based on the fact that the neonatal RF recipient is refractory to the induction of granulocytic leukemia by radiation (Cancer Res. 18: 842-848, 1958). This raises another question. Is this really the result of host variables or are we dealing with epidemiologic factors? We have to consider the probability of contact between the host and the virus which is encountered in the environment. Might these be an influence of passive immunologic resistance acquired through maternal antibodies? The epidemiology of the tumor-inducing material is, thus, another problem about which little has been said but which is certain to plague us as we try to reproduce results from one time to the next and with animals of different colonies and different ages.

Dr. L. Gross (Veterans Administration Hospital, Bronx, N.Y.): Dr. Law mentioned that there is a question whether the virus etiology of AK or C58 leukemia is an established fact. The fact that filtrates, prepared from AK or C58 leukemic donors and injected into newborn C3H mice, reproduced faithfully the same disease is now confirmed in many laboratories. Dr. Law felt that this would only indicate that a leukemic agent from AK leukemic tissues could be extracted. In certain instances too much criticism may create confusion, because under such conditions we could not speak of Shope rabbit papilloma virus or of Rous sarcoma virus. Everybody knows that a filterable agent from such tumors as the Shope rabbit papilloma, the Rous chicken sarcoma, or mouse mammary carcinoma can be extracted. Upon inoculation into susceptible animals, these agents reproduce the respective neoplasms. Accordingly, it is assumed that each of these three different viruses is the causative agent of the neoplasm from which it can be consistently recovered and which it reproduces after inoculation into a susceptible host. The same applies to mouse leukemia.

I am satisfied that AK leukemia in mice of the AK and C58 strains is caused by a virus and I assume that such a virus is transmitted from one generation to another because from normal AK or C58 embryos a leukemic agent could be recovered.

Many new problems are facing us, however, which are difficult to solve. We have carried out experiments similar to Dr. Kaplan's, as far as radiation-induced leukemia is concerned, with C3H mice which in our laboratory do not develop spontaneous leukemia except in rare instances. We induced some 50 to 60 percent leukemias in C3H mice by fractionated total-body X irradiation. This leukemia could then be passed by filtrates. At first the incidence was low, but, by selecting an active filtrate and passing it through newborn mice, we eventually obtained an active agent designated "passage X" which induces an incidence of up to 80 percent of lymphatic leukemia within only 3 to 4 months after inoculation into newborn C3H mice.

Is passage X different from passage A which we previously isolated from spontaneous AK leukemia? What is the relation between passage X, passage A, or any AK leukemic filtrate and Dr. Moloney's leukemia? I am not sure that the routine morphological methods will give a prompt answer to these questions. For instance, the passage A virus in our current experiments can be inactivated after heating at only 50° C. for 1/2 hour. Is this also true for the other leukemic viruses? Thymectomy will make C3H mice completely resistant to the passage of strain A virus. Thymectomy, however, will not affect the susceptibility of mice to the Friend virus, and apparently will have little effect on susceptibility to the Moloney virus. These may be some of the tools useful in attempts to differentiate these viruses.

The main problem we are facing now is: Do we have a single leukemogenic virus in our mice, or do we have a family of leukemogenic viruses?

Dr. E. A. McCuIIough (University of Toronto): I would like to ask Dr. Law a question. Some years ago he was coauthor of a paper in which he reported the injection of bone marrow into a cross between strains C3H and AK in which, if parental C3H marrow was used, the incidence in the F1 hybrid was reduced, and, if AK marrow was used, the incidence in the F1 hybrid was markedly increased. [Lorenz, E., Law, L. W., and Congdon, C. C: The role of bone marrow and spleen in induced and spontaneous leukaemia. Leukaemia Research. In Ciba Foundation Symposium (Wolstenholme, G. E. W., and Cameron, C. M., eds.). London, 1954, pp. 189-195.] I wonder if now, in view of Dr. Kaplan's results and in view of the importance of virus in leukemogen-esis, he feels that in this situation the bone marrow was acting as an antiviral agent. For the sake of argument, it might be assumed that the virus comes from the AK mice. Does this make the action of bone marrow a little more general than Dr. Law would allow Dr. Kaplan to have it?