1

Ehrlich was the first to observe the very interesting phenomenon that mouse-tumours, if transplanted into rats, grew well for about eight days, but that after this period the growth diminished, until finally the whole was absorbed. The occurrence of the same phenomenon was later described also for Jensen's tumour from the Laboratory of the Imperial Cancer Research Fund. Ehrlich claimed that this phenomenon was due to the exceptional virulence of his tumours, and could not be obtained for Jensen's tumour. Rats treated in this manner were further found to be resistant to all subsequent inoculations, a fact which he ascribed to the production of an active immunity in the animals.

When one had completed a long series of observations on what occurred to tumour grafts in immune mice, the investigation was extended to the observation of what occurred in rats immunised against mouse-tumour. A large series of rats was inoculated on the back with 0.1 c.c. of an emulsion of tumour 27. The rats were carefully observed from day to day, but no palpable increase in size of the mass injected could be recorded; instead, there was a gradual diminution until about the 7th or 8th day, when all trace of the injection had disappeared, or it was at least no longer palpable. This, as will be seen later, was not due to the absence of growth, but to the slowness of growth.

Fourteen days after the inoculation of the mouse-tumour, these rats were used for early stages of another 27 tumour, a control in normal rats being made at the same time. On examining the early stages in the control, pictures quite similar to those seen in normal mice were obtained, the rat-tissues proliferated and supplied a new stroma and vascular apparatus for the growing nodule. If any difference existed, the vascularisation ran a quicker course than in mice, and the amount of reaction on the part of the host-cells was greater. The epithelial elements divided actively, the mitoses were normal, and the development of new acini resulted in the formation of a perfect adeno-carcinoma similar to the parent tumour. Fig. 9 is taken from a nine days. old graft in a normal rat, and shows how at this date, the tumour-cells were still dividing. It will be seen that many of the tumour-cells have taken up the stain more than others, and are of a much deeper tint. This is, as was shown by Bashford, Murray, and Bowen, evidence of impending degeneration. Mitotic figures can also be seen in the stroma. of the graft depicted, and, indeed, it was most striking to see the 1 Carcinom Studien an Mäusen, 1905.

2 Proc. Royal Society, vol. B, 78.

manner in which these tumour-cells of the mouse were able for a time to make the rat-tissues subservient to their needs. As pointed out above, tumour 27 is a tumour with poor powers of growth-far behind those of Jensen's tumour. Ehrlich's explanation of the transitory growth of mouse-tumours in rats as due to the exceptional virulence of his tumours is shown to be untenable by the present observations, and those previously made with Jensen's and other tumours.

In the rats previously rendered immune, there took place a rapid destruction of all the elements introduced, so that as early as the third day, there only remained a few scattered tumour-cells, all of which showed marked signs of advancing necrosis (fig. 10). On the fourth day no formed elements of the introduced graft could be detected, but there was an active invasion of polyblasts and fibroblasts from the hosttissues, and the late stages of cleaning up of the field ran a very rapid course. They were accompanied by an active proliferation of the host connective-tissue cells over an extensive area. The same effect is obtained in rats when immunised with normal mouse-tissues, but not when rat-tissues are used.

Before passing to a general consideration of the results obtained, attention may first be drawn to the conclusions arrived at from detailed investigation of the changes in normal animals only. As already observed, Jensen, from the examination of early stages, was able to state that the new tumour arose from the epithelial elements introduced. Bashford and Murray later confirmed this result, and were able to add that the new stroma was derived by proliferation of the connective-tissues of the new host, and that the old stroma degenerated and became finally absorbed. The character of the stroma of a propagable tumour, with the exception of those cases where a sarcomatous transformation has occurred, remains constant within narrow limits. for the succeeding generations. Bashford, Murray, and Cramer stated that this can only be ascribed to certain properties inherent in the parenchyma-cells and influencing the reacting tissues of the host in a specific manner, for these are the only elements which continue from one generation to another. An alveolar carcinoma continues on transplantation to retain the delicate stroma of the parent tumour ; a hæmorrhagic adeno-carcinoma continues to develop large dilated vascular sinuses, and so on.

Ehrlich, in accepting this specific stroma reaction, likewise ascribes to the tumour-cells specific chemiotactic properties acting on the * Zeitsch. f. Krebsforschung, Band v. Heft. i. 1907.

FIG. 10.-27 19 J-20 D. Entire graft preserved 3 days after inoculation into a rat previously treated with 0.1 c.c. of emulsion of tumour 27. The whole graft is necrotic, even the superficial parts in contact with the tissues of the rat. tumour in the form of collapsed cysts with shrunken cells. Very abundant small cell infiltration Compare fig. 9, graft of tumour 27 in normal rat and fig. 4, a similar graft in an immune mouse.

A few islands of mouse
of surrounding tissues.
100
X 1.

fibroblasts of the new host. From the examination of a transplantable chondroma which very early showed hæmorrhages on transplantation, he came to the conclusion that the cells of this tumour possessed special angiotactic powers, which led to an over-production of capillaries. As mentioned by Gierke on an earlier page, it seems therefore possible to subdivide up the stroma reaction into two components, the fibroblastic, which leads to the development of the true supporting tissue, and the angioplastic, which provides the new blood-vessels.

Fig. 8, from a three day old hæmorrhagic tumour, serves as a good example of the case where the tumour-cells possess to an unusual degree an angioplastic chemiotaxis; whilst the figure of the stroma formation at four days, in tumour 27 (fig. 3), shows a case where the fibroblastic reaction preponderates.

Turning next to the consideration of the processes occurring in grafts in immune animals, the outstanding feature in all of the tumours employed has been the total failure of the new host to supply a vascular stroma. The failure of this re-organisation quite explains all the histological pictures which we have seen in the various tumours in immune mice. In the solid tumours the failure of a vascular supply led to the necrosis of all the graft, excepting only the cells at the periphery, which obtained their nourishment by diffusion from the hosttissues. In tumour 27, where the cells had arranged themselves in two layers, the one adhering to the host-tissues, and the other to the necrotic mass of the graft, the latter layer must have received its nutrition by diffusion across the cystic space intervening between the outer and inner layers. The fact of this tumour being able to grow along the free surface of the host-tissues, does not appear to be of fundamental importance, it is merely the expression of the inherent tendency of the epithelial cells to spread along a free surface. The cells of two other adeno-carcinomata which were tested in immune mice, but which are not given in the preceding description, showed the same power of lining a free surface, although to a lesser degree. In the experiments done by Ribbert upon the transplantation of pieces of normal epidermis, he found that the epithelium introduced, bent round in its growth, and by fusion of the two free edges led to the formation of a small cyst.

Having demonstrated the failure of the stroma reaction in immune animals, the question arises, how is this brought about? The failure of the reaction can be ascribed to one of two factors; either the tissues of the animal have been altered in such a way by the process of immunisation, that they no longer react to the stimulus of the cancer