Analogous phenomena are observed throughout the whole tumor virus field. Many different kinds of factors determine whether or not a cell can respond to a virus which may have been in cells from the very beginning of the life of the individual. It is clear that the genetic constitution of the mouse is a critical basic factor. In addition, the painful investigations of many workers have revealed the unquestionable influence of hormones. Yet, except for the establishment of a limited number of qualitative principles, the results of such work have actually advanced the total problem but little toward the solution of the question of the intimate cell-hormone-virus relationships involved in the induction of malignant change. If judgment can be made from the experience with a few other virus tumors, such as, for example, those of the bird, it can be assumed with assurance that significant progress with the mouse mammary carcinoma will be dependent eventually on quantitative correlations of the respective influences of virus and hormones. The critical unknown thus far is the virus. In consequence, regardless of the cost or difficulty, further elucidation of the processes leading to the onset of mouse breast cancer will not be made until the agent can be quantitated physically and biologically.

Dr. Andervont: Since Dr. Beard has told us so eloquently that the problem of breast cancer in the mouse is very simple, I am calling on someone who I deem has had sufficient experience also to tell us how simple it is, Dr. John Bittner.

Dr. John J. Bittner (University of Minnesota): Well, this has been very, very interesting. For about 5 years, this gentleman has been telling us that mammary cancer is a very, very simple problem. A few years ago I asked Ray Bryan, who worked on the mammary-tumor agent, and published in 1944, why he did not come back and clear up the mess which Dr. Beard says we are in. And Ray said, "John, I would not live long enough to do so."

I think, after listening to Dr. Muhlbock, that everyone who has worked on mammary cancer in mice is in agreement as to the role of the agent, hormones, and susceptibility. The only possible difference of interpretation could be explained by a difference between the incidences of the different strains maintained in Europe and in this country. Otherwise, the interpretation he has presented would be similar to that of others who have worked on the disease.

He said that, if the agent from five different strains was introduced into one host strain, there were differences in the incidence; of the stocks which he tested, strain A had a very inactive agent.

In this country Andervont and Heston have compared strain A and the C3H strain. There was no difference in the incidence of tumors in the reciprocal F1 hybrids. When we, using our strains, compared the reciprocal cross between strain A X C3H, those with mothers from strain A had a higher incidence than those with C3H mothers. In our hands the agent from strain A was more active than the agent from strain C3H.

When we made reciprocal crosses between the A, C3H, and DBA strains, in the A X C3H cross the strain A agent was more active. In the C3H X DBA, the C3H agent was more active, and in this cross the CSH agent was more active than the A agent in the A X C3H cross. In strain A X DBA, the A agent was more active, but when we crossed the DBA and the C3H, the DBA agent was more active than the A agent in the A X DBA cross. So you can take your pick, for the activity of any agent will be influenced by the constitution of the host.

I think that Dr. Muhlbock might give more consideration to the role of the adrenal in the development of mammary cancer. We have confirmed his work on the pituitary implants by inducing tumors in agent-free mice in which we cannot demonstrate the agent. In certain of our strains we can induce mammary cancer by implanting three pituitaries. Dr. Muhlbock has been able to induce mammary cancer by implanting one pituitary.

Dr. Halberg has obtained evidence that in mice with pituitary implants there is an increase in the serum adrenal corticosterone. In other words, he can pick up a difference in hormonal level in mice, after the implantation of pituitaries, that can be detected within 3 1/2 months.

Now, supposedly the pituitary is under the control of the hypothalamus. If pituitaries and a hypothalamus are implanted at the same site the development of mammary cancer is inhibited, not only in a low cancer strain, but in mice with the agent.

If the pituitary is implanted on one side and the hypothalamus on the other side of the mouse, the growth of the mammary cancer is accelerated.

If mice are painted with methylcholanthrene, mammary cancer is induced in some agent-free strains. If mice of the same strain are used with the agent, and 95 percent will have mammary cancer as either virgins or breeders, the incidence is reduced by painting them with the same hydrocarbon. Here is a compound which will induce mammary cancer in agent-free mice or inhibit mammary cancer in agent-containing mice.

Personally, I would like to see Dr. Beard work with mammary cancer in mice. He talks about the increase in incidence with the administration of an increased number of virus particles. When mammary tumors are assayed, the same extract will produce many more tumors when a 10^-4 dilution rather than a 10^-1 dilution is used.

It is regrettable that this does not fall into line with the chicken tumor virus. All we are doing is reporting facts as we see them, not as we would like to have them. If theories would solve the cancer problem, it would have been solved a good many years ago, but there is still a great deal of work to be done. In 1947 I was a coauthor on a paper suggesting that the role of the agent in the induction of mammary cancer in mice was to alter the hormonal metabolism, and I still believe that is the case-that the hormones become the primary inciters for mammary cancer in mice. If some investigators say we have an agent in these tumors that we cannot demonstrate, I feel it is up to them to show us how to do it. They can certainly get mice from us; I am sure that all of us will supply them with mice and they can work with their own animals in their own laboratory. They may survive longer than Dr. Bryan and solve the problem.

Dr. G. Miroff (University of North Dakota): Many years ago in Dr. Bittner's laboratory he related these complex situations to me. I was convinced that there must be a simpler way to test for the agent than by the protracted bioassay.