Dr. J. A. Murray, F. R. S.: The investigations of the last 30 years have proved that the cells of the higher vertebrates, under appropriate conditions, are capable of unlimited proliferation, and it is of importance in discussing the problems of experimental carcinogenesis to remember this fact, and to have clear ideas of the characteristic features of the proliferation of new growths, including cancer. This is very necessary because nothing is commoner in the literature of the subject than loose statements that this or that agent confers on the cells powers of unlimited proliferation, which is nonsense, seeing they already possess these powers. The essential feature is the uncontrolled or autonomous character of the cellular proliferation, that is to say, the agencies which are effective in the body in limiting the rate and amount of growth and cell division are ineffective against true new growths. This is true of both classes of new growths, the benign as well as the malignant. Examples are seen in the fatty tumours, lipomata, which go on increasing in size in an emaciated individual and the uterine myomata which grow progressively even after the menopause while the ordinary uterine muscle is shrinking or quiescent. The malignant new growths, carcinoma and sarcoma, exhibit this feature still more clearly because many show a more rapid rate of growth. The other distinctive features of the malignant as contrasted with the benign new growths are differences in degree, rather than in kind, and their more perfect independence or autonomy, manifests itself in infiltrative progress, disorganizing and destroying the normal tissues encountered, by pressure from without, or occlusion and rupture of blood supply. The stretching and tearing of the walls of blood and lymph vessels opens the way for the entrance of smaller or larger aggregates of the parenchyma cells into the vessels and these, transferred to remote situations further exhibit their independence and adaptability to new surroundings, by the formation of secondary centres of growth, or metastases. It is these manifestations of neoplasia which render cancer so formidable a problem in treatment. These new proliferative conditions arise in limited localized foci and once they have reached a size sufficient for recognition further increase takes place only from the descendants of the already transformed cells without fresh accessions from the surrounding elements of the same kind. The original focus may be solitary, or several foci (all minute) may appear almost simultaneously, and soon fuse into one tumour. One type of cell only acquires these new properties and by its multiplication gives rise to the new formation, so that it is usually possible by microscopic examination to their the tissue of origin, even after the tumour has reached a great size. The growths of any one tissue do not reproduce the full characters of the differentiated tissue of origin to the same extent. They form a continuous series ranging from apparently perfect reproduction of the histology of the parent tissue to a condition in which no specific differentiation can be recognized at all. Much unnecessary ink has been split in devising suitable words to describe this "undifferentiated," "dedifferentiated," "anaplastic" or "embryonic" state, but what is worth emphasizing is the fact the degree to which it occurs is relatively fixed in any one new growth, and is maintained practically unaltered throughout its course. The rate of growth shows a similar uncorrelated series of gradations and there is much evidence to show that this also is (or at least may be), an initial, inherent, quality of the essential constituent parenchyma cells.