All neoplasms consist of two main elements-the cells and the stroma. As has already been stated, the cells are derived from the various types of epithelial cells, and, with the exception of the tumors derived from the sebaceous glands, remain remarkably true to type unless flattened out by pressure. We can easily recognize the basal-celled and the spinous-celled type, and, with slightly more difficulty, the cubocelled type. The cells from the appendages of the skin may usually be recognized, although, of course, the tumors originating from the hair follicles may very soon come to absolutely resemble the neoplasms originating from the surface epithelium. The cell type of the sebaceous glands usually changes so promptly that it is impossible to prove that a tumor had its origin in these glands. This change consists in a loss of the fat, after which the cell is difficult to recognize.

The criterion of malignancy is the breaking through of the basal membrane by the epithelial cells. This basal membrane normally forms the limit of the epithelial cells, thus separating them from the fibrous tissue of the corium. Invasion, once the membrane is ruptured, may take place in any one of several ways, all carefully studied by Unna. The cancer cells may invade practically en masse, they may form more or less solid alveoli, they may push forward in long alveoli, or they may branch out like the branches of a tree, or they may invade singly or in small groups. The mode of invasion depends on at least two factors-one the inherent power of reproduction of the cancer cells, and the other the resisting power of the invaded tissue.

*Adami: Principles of Pathology, New York, 1908, i. 9Hertzler: Treatise on Tumors, New York, 1912.

Characteristic of cancer cells are the atypical mitotic figures, a peculiarity so pronounced that some authors have been inclined to believe that the atypical power of reproduction was the cause of the disease. It is now usually believed, however, that irregular mitosis is an effect and not a cause.

The degenerations of cancer cells have been carefully studied by several investigators, including Gilchrist* and Unna. Both call particular attention to the close resemblance of some of these degenerative bodies to parasites, and warn against the error of mistaking them for causal organisms.

Unna states that the most common degeneration is the hyaline one, the so-called cancer pearls being one variety of this type of degeneration. The pearls are simply groups of cells arranged concentrically, and having a great affinity for acid dyes, due to the degenerative changes. Unna also describes eight forms of hyaline degeneration of the circumscribed variety. They are:

1. Round balls of a size varying from 15 to 20 microns in diameter, and in the interior of which there exist the remains of a nucleus, always lying in a cavity.

2. Irregular, long twisted bodies are fairly common.

3. Small hyaline bodies, having a diameter of from 7 to 15 microns, may lie.in a nuclear cavity, but close to the nucleus.

4. The bodies described in the last heading may become elongated.

5. Hyaline rods or tendrils are sometimes found.

6. There may exist capsulated structures in which there is a well-preserved nucleus, lying in a ring-formed cell that has undergone hyaline degeneration.

7. Flake or vesicle-like hyaline bodies with nuclear cavities may occur.

8. And, lastly, there may exist hyaline vesicles in which the nuclear remains are inclosed by two hyaline capsules.

There is a final group in which exist conglomerations which have arisen from the compression and coalescence of the individual elements. There is another group of hyaline degeneration, even commoner than the circumscribed changes, and that is the diffuse hyaline change. This degeneration can be distinguished only in wellstained specimens, and is frequently a precursor of the more circumscribed changes. Unna states "the limited, diffused hyaline degeneration of the inner perinuclear substance of the cell-body is extraordinarily common".

*Gilchrist: Johns Hopkins Hospital Reports; Studies in Dermatology, 1896.

Next in frequency to the hyaline degeneration is the simple softening and liquefaction of the epithelium, which is often combined with the hyaline changes. A single cell may be attacked, or a large portion of the cancer may soften. This form of degeneration is not present in those forms of cancer that grow by long, slender processes.

A few cases show keratoid degenerations; that is, the formation of true horny pearls, so named because of their gross appearance under a lens of low magnifying power. This form of degeneration occurs solely in tumors derived from the prickle layer. The cells are arranged concentrically in whorls, and are apparently under considerable pressure. Naturally they have a great affinity for acid dyes, such as eosin or orcein.

Occasionally calcification occurs in cancer, developing from the deposits of lime salts in the pearls.

The stroma or fibrous tissue is either the old fibrous tissue, which has been invaded by the tumor cells, or is newly formed. In some types of neoplasm one prevails, while in other types of tumor the other prevails. Of course there is frequently a combination of the two. Quite frequently the cancerous invasion of the corium induces a severe inflammatory reaction, some authors going so far as to claim that a cancerous downgrowth of tissue never occurs without a cellular exudate forming around it. The cells usually found are small round cells, fixed tissue cells, mast cells and plasma cells. Giant cells are occasionally found. Sometimes this reaction is so marked as to almost completely obscure the cancerous process. Unna states that in twenty-four out of his seventy cases an inflammatory reaction was very considerable; the rodent ulcers, Paget's cancers, and the cancers of xeroderma pigmentosum in particular showing much infiltration. It is very rare to find every nodule or process of invading cancer walled in by an inflammatory membrane, but some of them usually are. Unna states that in ten of his cases there was no inflammatory reaction. The cellular infiltration of the cutis is usually directly in ratio with the rapidity of infiltration by the neoplasm.

Neoplasms have a complete vascular circulation, the blood vessels may pre-exist, or may be newly formed, but are usually derived from both sources. The newly-formed blood vessels consist of an endothelial lining, with occasionally a fibrous outer layer, and vary greatly in size.

According to Hertzler neither nerves nor lymphatics have been satisfactorily demonstrated.