Dr. Law: I do not quite understand your question, but I think that both Dr. Kaplan and Dr. Gross missed the point of my first comment. I said that, in my opinion, the present knowledge in this problem concerning AK and C58 leukemias is that there is obtainable from AK mice an agent which is leukemogenic. But it seems to me, when you say that this agent is responsible for the spontaneous leukemias in AK and C58 strains, it is still a concept, because there is no proof as far as I know. The explanation is in the form of a concept and must be proved. Why have strains AK and C58 for 20 or 30 years, generation after generation, shown high rates of lymphocytic neoplasms? What is the nature of the transfer of virus?

There is no doubt that these materials are leukemogenic in the recipient mice into which they are introduced, and I am not arguing that point at all. When they are considered responsible for the spontaneous leukemia of these strains, this is another question. I do not know that anybody has an answer as to how the bone marrow acts in the prevention of X-ray-induced leukemia.

Dr. Kaplan: I do not believe that we will get much further with this argument. It becomes a question of semantics whether something is an established fact if any one individual in the world does not believe it.

I would say that this is a problem relating to the philosophical nature of evidence on causality. I feel that Dr. Gross's agent is extracted from leukemia, that it will produce morphologically identical leukemia in other strains, and that it will accelerate the appearance of morphologically identical leukemia in the strain in which it was first encountered.

I believe that we will get further by advancing with the problem in the laboratory. Once we succeed in establishing in vitro systems, in which these extraordinarily bashful leukemia viruses can be studied more directly, we can clarify many currently troublesome problems. If we use up too much of our energy at the present time in arguments of this sort, I think we will hold back progress.

The radiation-induced lymphoma is interesting to compare with the spontaneous lymphoma. The spontaneous AK system apparently does not require any radiation, and one wonders why not. I think it would be interesting to go back to the AK tissues and see if there are any indications about this. On one point I will definitely agree with Dr. Law: We are still in the dark as to the pathogenesis of leukemia in AK mice even if it is conceded that they have a virus. It takes them months to get the disease, and we have very little understanding of what finally triggers this response.

There is a great need to return to the AK system and study it further. But, in the irradiated system with which we have worked, it appears that radiation is a leuke-mogen, whereas many other agents are not leukemogens. Radiation, unlike all the other agents we have studied, does two things, and both of these must apparently occur in order to produce leukemia: (1) damage to the thymus; I would guess that in the course of that injury virus is released from some thymus cells in which it has been confined; (2) damage to the bone marrow, which in turn interferes with the recovery of the thymus, and as I have shown, the nature of that interference seems to have something to do with a maturation arrest.

Cultivation of the Mouse Mammary Carcinoma Virus

The description in 1948 by Porter and Thompson (1) of particulate bodies associated with neoplastic mammary cells observed in electron micrographs furnished a strong argument in favor of the viral nature of the milk factor. However, two important links were still missing: (1) No correlation could be established between the particles observed and the carcinogenic activity of the milk, and (2) there was no satisfactory evidence that the agent could be isolated from the tumors and propagated independently in tissue culture.

Two years ago, Moore et at. (2) were able to show by physical procedures that the particulate bodies seen with the electron microscope are endowed with infectivity, and thus succeeded in identifying them with the milk agent. At the same time strong evidence was obtained that the mammary-tumor agent (MTA) multiplies in tissue culture (8).

Earlier attempts by Taylor and coworkers to maintain the milk agent in embryonated eggs had given controversial results. However, Bittner, Evans, and Green (7) recovered an active agent from the yolk 12 days after inoculating eggs with a cell-free tumor nitrate. Also, by injecting mice with an agent maintained through several passages on cultures of chicken fibroblasts, Pikovski (8) increased the tumor incidence in a colony in which it was usually low. More recently we observed that the particulate bodies released by the neoplastic cells could increase considerably in number simply after a few hours of cultivation in roller tubes (9). These were definite indications that the MTA can be maintained, possibly multiplied, outside its natural host with conventional techniques of virus production.