cases, so that a prognosis is specially uncertain in the case of this growth. Local recurrence may in almost all cases be predicted where the knife has not gone quite clear of the affected tissues; for this form of tumour, usually described as encapsuled, or spreading only in the connective tissue, occasionally exhibits distinct infiltration of other textures. I have myself seen instances of spindle-cell growth invading muscular tissue, and breaking up the striped fibres in quite as destructive a manner as any carcinoma (see Plate I, fig. 4). Infection of remote parts is also not very uncommon. I have met with spindle-cell growths in the liver, lungs, and mesentery; but such secondary growths are very rarely found in the lymphatic glands.

It is probable that a careful consideration of the microscopic structure offers valuable indications for a prognosis, those tumours being more apt to show true malignancy whose elements are plump and interspersed with large, irregular, and multi-nucleated cells. Special regard must, of course, be had to the rapidity of the growth, its consistence, and its seat. Where, for instance, the tumour is distinctly encapsuled, and even pedunculated-as in certain fleshy nasal polypi-the prognosis would be infinitely more favourable than where the growth infiltrates a soft, moist part, subjected to constant movement as I have twice seen it in the uterus, simulating, in all respects, ordinary uterine cancer, and running a similarly fatal course.

CHAPTER VI.

SARCOMA (continued).-Round- and Oval-cell Sarcoma-Small Round-cell Sarcoma-Myeloid Sarcoma.

ROUND- AND OVAL-CELL SARCOMA-Besides the spindle-cell growth which constitutes the bulk of most of the sarcomata, certain of these tumours are made up of cells of a still lower type-embryonic cells-consisting merely of little lumps of nucleated protoplasm, which, held together by a nearly fluid connecting-material, show but little tendency to assume an elongated form by regular pressure on the sides, but remain as soft, spherical or ovoid cells, like those met with in granulations, although occasionally considerably larger; hence the name "granulation tumour," which has been suggested for this form of growth. Tumours possessing this structure are generally very soft, and from their consistence and general appearance, as well as from their microscopic characters, are more liable than any other form of neoplasm to be mistaken for medullary carcinoma. These tumours are very vascular, the vessels being usually simple channels through the celltissue, with walls formed by the same cells; and hence the very frequent rupture of the vessels, and escape of blood into the substance of the tumour-sometimes to such an extent as to give rise to a large "blood-cyst," which by its size may mask the real nature of its origin. The cells themselves are also very fragile, so that it is unusual to prepare either a scraping or a section of one of these growths without allowing a large number of free nuclei to escape. Seen in section, it is often even more difficult to recognise the precise

form of the cells than in the fusiform variety. The usual appearance is that of an amorphous or granular substance in which large or small round or oval nuclei, with bright nucleoli, are imbedded. At the edges of the section the rounded outline of the soft cells may be recognised, and in little detached bits of the section their shape becomes yet more distinct. The size and shape of such detached cells (see fig. 14) suggest at once the occasional difficulty in distinguishing these tumours from carcinoma.

The cells are, indeed, somewhat plumper and less like epithelial elements than are the ordinary cells of carcinoma, and there is usually some amount of visible intercellular substance; but where the stroma is fibrillated-as is sometimes the case-and fibrillated in such a way as to suggest a meshwork like the alveolar stroma of carcinoma, the distinction between the two is one of the most delicate points in pathology. Billroth has suggested the name "alveolar sarcoma" for this latter form. It is, fortunately, rarely encountered, and its accurate distinction is a matter of patho

logical interest rather than of clinical importance, for about its grave malignancy there can be no doubt. I have seen a patient dying with upwards of a hundred of such tumours scattered through her body, chiefly in the subcutaneous cellular tissue, but also affecting lymphatic glands, breast, and kidney. In this case most of the growths contained more or less of black pigment, the pigment granules lying in large round cells, and in sufficient quantity to render the tumours distinctly melanotic.

The more usual form of round-cell sarcoma is made up of much smaller cells than those just described, cells closely resembling leucocytes or the first cells of a granulation; and since these minute cells are mostly about the size of bloodor lymph-corpuscles, the distinction between this form of growth and lymphoma (to be presently noticed) is sometimes as difficult to draw as is the distinction between the larger oval-cell sarcoma and medullary carcinoma. This small round-cell tumour differs from true lymphoma mainly in lacking the fine reticulated stroma of the latter neoplasm. The cells or corpuscles are separated by a variable amount of semi-fluid granular substance; but occasionally an appearance as of fine fibrillæ branching amongst the cells is met with, and when this is the case the growth is to be distinguished from lymphoma by the absence of minute nuclei in the angles of the network to be hereafter described as characteristic of that structure. Round-cell sarcoma is a distinctly infiltrating growth, as may be seen in the subjoined sketch (fig. 15), taken from a thin section of one of these tumours springing from the fibula and blending with the muscles of the calf. Remains of striped muscle-fibres broken up and invaded by the new growth sufficiently attest its destructive character, and show that the muscle is not merely wasted by the pressure of the increasing mass in its neighbourhood.

The structure of this tumour closely resembles what is to be met with in the vicinity of nearly all active new growths. As these advance they send before them, so to say, "feelers"

of this "indifferent granulation material," which stretch out into the surrounding structures, and form the first histological

FIG. 15.-Small round-cell sarcoma infiltrating muscle; a a a, nuclei of the striped muscle-fibres. Magnified 220 times.

indication of the march of the morbid change (see Plate II, fig. 1). Reference has been made to this tissue in the description of the early stages of carcinoma. The peculiarity about the growth now occupying our attention is, that the whole bulk of the possibly enormous mass is made up of the same simple small corpuscular structure. To establish this fact it is of course necessary to take a scraping or section from three or four different parts of the tumour under examination. Such an investigation may show the prevailing type of cell to be of another kind, these small round corpuscles merely spreading about the margins of the growth; and so the tumour may receive a different name, and perchance a corresponding difference may be required in the prognosis of the case.

These round- and oval-cell sarcomas are usually very soft, white-or variously mottled by the results of blood-extravasations-blending intimately with the structures amongst which they lie, and readily exuding a creamy juice filled with