Cavern- Carcnioits Blood-tumours. The pigmental alveolar sarcoma is ous sufficiently common in the situation of congenital mother-marks tumour, of the skin to be one of their characteristic developments. Another of their developments or equivalents is the nsevus or angcioma or cavernous tumour, whose structure may be said to consist, iu general terms, of a spongy mesh work of alveolar spaces, bounded by coarse and elastic trabeculffi and filled with blood. Arteries open into such tumours and veins pass out from them, the cavernous territory being intermediate ; but, according to several authorities, this con nexion with the circulation is not primary to the cavernous tumour but acquired. "Without entering upon a discussion of details, the analogy of the alveolar sarcoma growing on the same basis of a congenital pigment-spot may be kept in view. The^alveolation is the same in both cases, although the trabeculre in the cavernous tumour are somewhat stouter, the grand difference being in the contents. If, however, we suppose the epithelial-like cells of the alveolar sarcoma all to become large and filled with a yellowish colouring matter, as indeed many of them do, and if we suppose that these honnatoblasts (for such they are) go on to fulfil their destiny, then we should have a cavernous blood-tumour, that is to say, the alveoli would be filled with red blood-corpuscles. It will not be possible to offer evidence of this process except for the cavernous blood-tumour of the liver, an organ in which such tumours are comparatively frequent, and mostly in later life. The cylinders of liver-cells appear to become nar rower and narrower, as if from pres sure of the capillaries, and ultimately to disappear. From the supporting tissue a new growth of cells takes place (fig. 28). These are hsemato- blasts ; their protoplasm becomes red blood-disks, and their nucleus sur vives with the remarkable trefoil arrangement of cleavage which has been described for several other in stances of the hpematoblastic pro cess. There can be no mistaking the identity of this process with that of the blood-cyst of the neck already FlG 28.-Ha^atobla8tic process in mentioned ; it is essentially a main- cavernous growths of liver (dog). festation of hrematoblastic function late in life, differing from that of the blood -cyst in the fact that the centres of blood-formation are sepa rated from one another within alveolar boundaries. These cases illustrate another striking property of cavernous blood -tumours, namely, to heal spontaneously in parts or to develop embryonic scar- tissue through more or less of their extent (fig. 29, a). The ordinary cavernous texture of an angeioma is produced by the formative pro cess stopping short of embryonic connective tissue or scar-tissue. The accompanying figure (fig. 29, b) is from an enormous angeiomatous a, the supporting tissue producing hiematoblasts b ; c, nuclear re mains of hsematoblasts and red blood-disks side by side. Fio. 29. a, cicatricial tissue from cavernous tumour of liver (dog) ; b, mesh- work occupied by red blood-disks, from cavernous tumour of head (ox). tumour on the side of an ox s head ; the structure is very like that of the young connective tissue of the former figure, except that the meshes are densely packed with red blood-corpuscles. There are, however, other parts of the tumour where the fibres are broader, the meshes narrower, and with embryonic cells lying in them, in stead of or along with blood -corpuscles. There is no definite limit between such cavernous blood-tumours and true blood-cysts ; in the latter the numerous hfematoblastic centres open communications, and the further process takes place in the cellular tissue forming the cyst-wall. Traces of The blood-making office of the mesoblast is the earliest and blood- greatest of the functions of embryonic cells, and it is not surprising making that it should come out more or less obviously in those formative in processes in the common binding tissue of the body where there is a tumours, persistence or revival of embryonic activity. We seem to find traces of it in the pigmentation, in the cystic excavation, in the alveolation, in the mucous or myxomatous transformation, and in the cavernous structure of mesoblastic new growths. The embryonic spontaneity in the middle layer is, of course, wider than mere blood-making ; but the hfematoblastic function or tendency is certainly the most fundamental, and the traces of it in the foregoing tumours are our best help towards a rational interpretation of them. Persisting or revived embryonic activity in subcutaneous and other homologous tissues cannot but bring to light more or less of this all-important mesoblastic function ; the memory of it is too strong to be ignored. We come next to a function of embryonic cells which is only second to the h.Tinatoblastic, namely, the osteoblastic or bone-making func tion ; and even with the bone -making process the earlier blood- making process is deeply interwoven, for in the marrow of the bones the hsematoblastic activity of cells persists long after it has ceased elsewhere. The bone -making function of embryonic tissue if function it Turm may be called comes into a large number of tumours ; or, in other of bo: words, a large proportion of all mesoblastic tumours are tumours of the bones. In all of these the embryonic law of development and growth is clearly present. The results, however, are frequently more complex than in the tumours hitherto considered ; or, in other words, tumours of the bones are exceedingly liable to have a structure so mixed as almost to baffle systematic description. One reason of this is that the osteoblastic and hsematoblastic functions of embryonic cells go hand in hand in their production ; and the complexity of structure is, accordingly, greatest in those which grow from that part of the bone where the blood-making resides, namely, the marrow. The other great formative tissue of bone is the periosteum, a tissue which retains its embryonic struc tural features long after the mesoblastic tissues elsewhere in the body have lost theirs. The marrow and the periosteum are fre quently involved in the same tumour ; or an essentially similar morbid product may be derived from either. That is notably the case with the tumours of the bones which we come to first, the cartilaginous tumours or enchondromata. Ecchondrosis. It is only rarely that a cartilaginous tumour Carti grows from cartilage, the observed instances having occurred at the tumo cartilaginous lines of union of the base of the skull, at the epiphy- sial lines in long bones, and in such permanent cartilages as those of the larynx and trachea. To these direct outgrowths of cartilage- cells Yirchow has given the distinctive name of ccchondroscs. Usually the cartilaginous tumours do not grow from pre-existing cartilage ; they grow either from the periosteum or the marrow of the bones, or they form in certain glandular organs, especially the salivary glands (parotid, labial, &c.), the mammary gland (oftenest in the dog), the lacrymal gland, the testis, &c. These latter enchondromata are a class apart, involving considerations of disordered everyday secretion rather than of the revival of embryonic activity (see "Errors of Secretion," p. 379 below). The enchondromata that fall to be considered here are those which grow within or upon the mctacarpal bones and the finger-bones, more rarely in the corre sponding bones of the foot, not unfrequently in the bones of the face, and, it may be, in the leg-bones and arm -bones, or in bone anywhere. Enchondroma. The simplest cases (but the least frequent) are those that form between the periosteum and the hard bone from the growth and transformation of the cells of the periosteum, being directly homologous to the ensheathing cartilage-callus of repair. They differ from the cartilage of repair in precisely the same way that a granulation-like sarcoma differs from the granulation-tissue of repair, that is to say, the existence of the tissue is not self- limited, or it has no tendency, or only a feeble tendency, to cica tricial modification, shrinkage, or absorption. These purely sub- periosteal enchondromata are said by Paget to be nearly character istic of the ends of long bones, although they do not encroach on the articular cartilage. When a cartilaginous tumour occurs in the shaft of the bone it is partly subperiosteal and partly in the marrow ; and in the most characteristic seat of enchondromata, the bones of the fingers, the growth is entirely in the marrow if the tumours are multiple ; but, curiously enough, it is subperiosteal if there is only a single tumour (Paget). There are also cases where islands of cartilage form in the compact substance of the bones, corresponding to Haversian systems. The tissue-affinities of a cartilaginous tumour growing between the periosteum and the hard bone are not difficult ; the homologue, as we have said, is the callus-cartilage of repair. The histogcnesis and physiological analogies of an enchondroma of the medullary canal of a bone are less easy. We know that the marrow was pre ceded, in the development, by a bluish rod of fa-tal cartilage, of which all characteristic traces had disappeared before birth. As the blood-vessels entered it, it had changed into a spongy kind of bone, in whose spaces lay many spherical nucleated cells retaining a hsematoblastic or blood-making function ; all the spongy bone is gradually absorbed in the shaft, the last traces of it being a few spiculrc on the hard inner wall of the medullary canal, and the cavity is occupied by a highly vascular substance, the red marrow characteristic of young bones. The spherical cells of the red mar row become excavated into fat-cells, and the red colour changes to yellow. It is probably in this final phase of the development inside the shaft of a bone that we must look for the opportunity of the central enchondromata forming. The secret of the return to cartilage in some cases, and at certain spots, probably lies in the change of red marrow into yellow ; instead of becoming fat, it
