to the presence of mucus, but in the form of a degeneration; and a colloid tumour bears to a myxoma much the same relation that a fattily degenerated fibrous tumour does to a lipoma.

Thus it is that, besides such very characteristic examples of colloid cancer as have been best described by Mr. Sibley, in his admirable paper in the 'Medico-Chirurgical Transactions,' we meet with colloid nodules in many examples of carcinoma and sarcoma, and also in cases of epithelioma; and even single cells are sometimes the seat of this change, appearing amongst the polymorphous carcinoma cells, with large refracting clear spaces within them, distending the whole cell, or giving the impression of a greatly dilated nucleolus. Where this change is general, a peculiar jellylike aspect is imparted to the growth, which can yet be seen to be a genuine infiltration. Thus, I have seen a portion of a very extensive colloid carcinoma in the abdomen (its usual seat) distinctly infiltrating the muscular tissue of the uterus, growing from the peritoneal surface, and invading the organ to a depth of some lines. The effect of this accumulation of mucus within and between the cells is to alter the microscopic appearances as greatly as the coarser characters. Single oval cells become enormously distended, the nucleus remaining near the periphery, and the rest of the swollen cell being marked with faint concentric lines (possibly indicating successive stages in the mucous accumulation), and in this way a figure resembling an oyster-shell is produced (fig. 10). When a multi-nucleated cell is affected, the group of nuclei is surrounded by the same crease-like lines, and when, from the over-distension of many cells, a distribution of the contained mucus through the stretched alveoli takes place, large spherical spaces are formed, in which the remains of cells and nuclei in the centre are surrounded by these singular concentric faint lines, so that a microscopic portion of carcinoma so affected reminds one strongly of a sheet of still water, into which a few pebbles have been tossed, throwing the calm surface into a series of circling ripples, which gradually fade

into one another, and are lost (fig. 10). The fibrous stroma, however, may remain comparatively unaffected, although much stretched, and broad tracts of waving fibres with oat

FIG. 10.

9

FIG. 10.-Colloid carcinoma.

The two detached cells exhibit in a marked

degree the change which has affected to a less extent those still grouped in an alveolus. From a colloid of the stomach. Magnified 220 times.

shaped nuclei may be traced mapping out the growth into round or oval alveoli of various dimensions (fig. 11). These appearances have gained for the tumour the name "alveolar cancer;" but it has been already shown that all carcinoma is essentially alveolar. It is only that here, the alveoli being distended with a clear fluid, their arrangement is more clearly discernible.

It is to be noted also that in some cases of colloid carcinoma the cells themselves are quite free from any but the ordinary fatty degeneration. The viscid mucoid fluid appears to be poured into the alveoli from some other source, leaving the unaltered cells squeezed into clusters in the centre of the greatly enlarged spaces. I have lately seen two beautiful

in stances of this change affecting ordinary scirrhus of the breast in the practice of my colleague Mr. Croft, and in both cases one could trace the gradual increase in the size and

FIG. 11.

FIG. 11.-Colloid carcinoma, showing the fibrous stroma, whose spaces are filled with clear mucus. At one point, a few cells undergoing the colloid degeneration are shown. Magnified 220 times.

shape of the carcinomatous alveoli by the accumulation of the mucus within them until the stroma was almost entirely replaced by large tracts of the fluid in which floated the cellclusters mentioned.

MYXOMA, the growth with which these colloid degenerations are so often confounded, is a tumour of mucous tissue -the tissue which occurs so abundantly in the fœtus, but which is met with elsewhere only in the vitreous humour of the eye and in the jelly-like basis material of the umbilical cord. Similar as are the coarser characters of these tumours to those of colloid cancers, the microscopic structure is very different. Myxoma in its purest form resembles a lump of jelly limited by a very thin capsule, and separated into lobules of varying sizes by delicate curving and interlacing

fibres. When cut into, the knife drags out of the tumour long viscid strings of clear fluid, and the same fluid wells freely from the cut surface, and flows away in a bulky stream. This fluid gives the characteristic reactions of mucus; that is to say, it is coagulable by heat and vegetable acids, and forms a precipitate with mineral acids, soluble in an excess of the acid. Its mode of coagulation in spirit is also peculiar, for instead of forming dense clouds or flocculi, as albuminous fluids do, its fall through the spirit is marked by circling streaks and slender ribands of opacity. A little bit of the fibrous stroma teased out or pressed between glasses shows a very definite structure, cells of either a spindle or irregular shape branching off into long, slender, communicating fibrils, which by their interlacing form an extremely delicate spongy tissue, pervading the tumour and holding the mucus in its meshes. Such a cell structure is shown in Plate III, fig. 4, that drawing being taken from a pure myxoma the size of a cocoa-nut, lying between the muscles of the thigh of an elderly woman. In figs. 1 and 2 in the same plate other varieties of the same tissue are represented, taken from a soft complex tumour of the parotid. Myxoma is very rarely malignant, and hence it is important to distinguish it from colloid degenerations of such formidable growths as sarcoma and carcinoma.

It happened to me, for instance, to meet with a case of pure myxoma of the breast, which had been removed for cancer, at the same time that I was engaged in the examination of the two breast specimens of colloid just mentioned, and the coarse characters of all three tumours were so very similar that it would have been quite impossible to pronounce this one to be a myxoma before a microscopical investigation revealed its structure of beautfully delicate interlacing spindle cells.

CHAPTER V.

SARCOMA.-Distinction from Carcinoma-Divided into Spindle-cell, Roundand Oval-cell, and Myeloid Sarcoma, and Glioma Sarcoma.

Spindle-cell

IN the last chapter those varieties of carcinoma which have acquired dintinct names, as fungus hæmatodes, colloid cancer, and the like, were discussed, but a large number of malignant growths yet remain for consideration, which in their intimate structure are essentially distinct from carcinoma, although generally coupled with it by English surgeons under the comprehensive title cancer. The most important of these may be divided into two groups, the sarcomata and the epitheliomata; and of these sarcoma claims our first attention, as the more malignant, and, from a clinical point of view, the more nearly allied to carcinoma. It is probable that by far the greater number of "soft cancers" and "firm medullary cancers" recorded in English publications are really sarcomas; and if we were to limit the term "cancer" to "carcinoma," as here anatomically defined, we should probably be surprised to find how very rarely true primary medullary cancer is met with by the surgeon. Nearly all melanotic growths, for instance, and all the primary soft cancers of bone, would probably be more correctly classed with the sarcomata; a fair proportion of "villous cancers," and at least a third of the cancers affecting the uterus and rectum would find a place amongst the epitheliomata, whilst all the "medullary cancers" of the eyeball in young children are really gliomas. Allusion has been already made to the clinical importance of recognising these distinctions, and it will be sufficient to repeat here