diseases, and notably those in which there is a tendency to heterogenous proliferation of epithelial cells. Wickham, Darnier, Hacke, Malassez and Albarran, Steinhaus, Strobe, Vincent, Foa, Van Henkelom and Thoma, have all reported such parasitic cells in carcinoma, and the first two observers have regarded them as the probable cause of the disease though they cannot be demonstrated as etiological factors, nor have cultures or inoculations been made. On the other hand, Russel, Pifford, Schutz, Cornil, Hausemann, Törok, Ribbert, Tommasoli, Duplay, Cazin and others, who have observed these bodies, incline to regard them not as parasites, but as products of cell degeneration or as karyokinetic nuclear changes.

Within the past year three works on the subject, by Podwyssozki and Santschenko, Sondakewitsch and Noeggerath, have appeared. The first two speak with great assurance on the presence of parasites in carcinomatous cells; though after their exhaustive research, they conclude that neither the observers, who claim the presence of sporozoa in carcinoma, nor those who deny their presence, have sufficient ground for their dogmatic claims. Stroebe, who has observed and delineated these bodies, does not feel convinced of their parasitic nature, because of the difficulty of exeluding the question of nuclear degeneration.

Observations have also been made in the line of the discovery of bacteria in carcinomatous growths. Rappin claimed to have discovered a diplococcus as the cause of carcinoma, and even to have inoculated rabbits with positive results. Scheurlen found a spore-forming "carcinoma bacillus," which he was able to cultivate on gelatine and potatoes. But these observations have been long since thrown aside, either on the ground of inaccuracy or the want of further proofs. Finally, we have the elaborate researches of Adamkiewicz, just published. The wide-spread attention which they have received, and the immediate clinical application to which they have been put, entitle them to a more complete analysis in this paper. Adamkiewicz holds that carcinoma presents all the characteristics of a chronic infectious disease, but not being able to discover the parasite of the disease, we have too readily fallen back upon the old hypothesis of Cohnheim. He refuses to recognize carcinoma cells as identical with epithelial cells, and points out that they present three forms in the course of their development. The young form is that resembling the leucocytes; in a higher stage of development they approach the epithelial cells; and, lastly, they become carcinoma cells, with their tendency to the formation of irregular projections, and finally to degeneration and disintegration. This degeneration in the later stages he regards as a feature of differentiation from the epithelial cells. He assumes that carcinomatous tissue possesses toxic properties, and proves it by inoculating animals with bits of tumor and by injecting filtrated watery extracts, producing death in a few hours or days. This special poison he has called cancroin. It seems, without doubt, to him that carcinoma is a parasitic discase. The parasite not growing on the usual culture media, he undertook the use of living tissue in its stead, and planted bits of carcinoma in the brains of rabbits.

The carcinoma cells were observed to migrate from the implanted particles in the direction of the least resistance along the clefts in the cells in the brain-substance, to become implanted and form nuclei for further propagation. Therefore, he concludes that the carcinoma cells are to be regarded as living creatures, and are themselves the specific parasites of the disease. To this parasite he has given

the name coccidium sarcolytes, and this is the source of the cancroin, the irritating action of which causes the changes in normal tissue.

When we now come to sum up these observations, we see the very conflicting opinions of accurate observers, and the diametrically opposite conclusions at which many have arrived. The theory of the bacteriological origin of carcinoma gave place, among those who sought for a parasitic etiology, to the sporozoa or coccidia. Against these stand the views of a great number of observers who believe that these bodies, which have been interpreted as animal parasites, are the products of disintegrated nuclei, or nuclei in the process of mytosis. The whole evidence of their parasitic nature is based upon the insufficient ground of their appearance within or among the carcinomatous cells. They have not been isolated or cultivated outside of these cells. The microscopic appearance alone is inadequate for the basing of such a conclusion. For example, I may cite the deception occurring in fat cells, by which the intracellular septa, when stained, present the appearance of tubercle bacilli, a thing which Orth has regarded of sufficient importance to be made the subject of an essay. The pronounced tendency of carcinoma cell nuclei to disintegrate, and the mytosis occurring in the younger cells may well account for these peculiar appearances.

The fact that certain granulation tumors are of infectious origin does not justify the conclusion that carcinoma, clinically and histogenetically different, has a similar cause. The transplantation or infection with carcinoma cells is nothing more than can be done with the skin, periosteum or other tissues, and occurs in the benign as well as malignant growths. With the granulation tumors it is different. A secondary growth of tubercular or syphilitic tissue is not brought about by the transplantation of the cells making up the growth, but by the transplantation of the specific microorganisms by the blood, lymph or other channels, the irritating products of which give rise to the peculiar productive inflammatory process. The cells of such a secondary deposit are by no means descendants of the cells of the primary growth as are those of carcinoma, which when implanted remote from the mother tumor, continue to produce cells of the parent type, the preexisting cells taking no part in the formation of the characteristic parenchyma. The only means by which secondary deposits can form from carcinoma is through this transplantation or metastasis of carcinoma cells, which is facilitated by the peculiar anatomical nature of the tumor. If, then, there are specific parasites for carcinoma, they must possess a predilection for epithelial cells, and be transmitted by the cells to the localities in which metastasis develop, and yet not invade the cells of this new locality, even though they be epithelial in character. The parenchyma of a metastasis in the liver of an epithelioma of the tongue presents the peculiarities of epithelial cells of the tongue. The flat epithelial cells proliferate among the polyhedral liver cells, which play only a passive part.

It has been my fortune to observe a primary carcinoma of the liver with multiple secondary deposits. On cutting into these secondary nodules, which were scattered along the post-peritoneal lymphatics, a greenish-yellow fluid exuded which responded to the tests for bile, and the cells of these nodules were anatomically identical with liver cells. Examples of this sort, showing that in the metastasis of carcinoma the parenchyma repeats the type of the epithelial cells from which the primary growth sprang, might be indefinitely multiplied.

These facts speak against the microbic theory of carcinoma, and quite refute the doctrine of Adamkiewicz, which asserts that the characteristic neoplasmic cells. themselves are not epithelial cells, but are parasitic bodies. How he accounts for these bile-secreting parasites is hard to conceive. Furthermore, it is not difficult to make a section at the edge of a carcinoma, including some of the normal tissue, in which the normal epithelial cells may be traced as directly anatomically continuous with the parenchymatous cells of the neoplasm. The same thing is seen in the adeno-carcinomata so common in the stomach and rectum, in which adenomatous cells merge into the carcinomatous cells.

If the parasitic theory be accepted, it must be reconciled with the other malignant tumors, and also with the so-called benign growths which give rise to metastases, and which have so much in common with carcinoma. The fact that the carcinomata are anatomically related through the endotheliomata and peritheliomata to the malignant sarcomata on the one side, and that they are allied through the adeno-carcinomata to the benign adenomata on the other side, places another difficulty in the way of this hypothesis. This similarity may be further traced in the etiological factors strikingly common to all tumors, which may be spoken of as mechanical or chemical irritation, or influences which hinder in some way the development of cells within their normal limitations. The epithelioma of the lower lip of pipe-smokers, the chimney-sweep's carcinoma scroti, the "paraffine cancer" of paraffine workers, carcinoma of retained testicle, epithelioma of the tongue opposite the sharp edge of a carious tooth, are but a few of the wellknown examples. In scar-tissue of the soft parts develop carcinomata or sarcomata; at the seat of fractures appear chondromata, osteomata or sarcomata; from chronically inflamed mucous membranes develop polypi, adenomata or carcinomata; and so on examples might be multiplied. The tendency of an heterogenous proliferation of epithelial cells at the seat of a lupoid inflammation gives a glimpse of bacterial products acting as a cause of epithelial misdevelopment. In the edge of a healed gastric ulcer the epithelium is observed to send down chains infiltrating the underlying tissue. As a rule this carcinomalike condition gives rise to no further growth, but when, for some reason, the physiological equipoise between the aggressive tendency of the epithelium and the resisting tendency of the surrounding cells becomes disturbed, from precisely this locality carcinoma develops.

Nor do the malignant tumors alone give rise to metastases, for multiple developments of chondromata, lipomata, fibromata, myomata, papillo-cystomata and in fact, all the new growths, benign or malignant, have been observed to proliferate in this manner.

It would be superfluous to multiply evidence showing the etiological and anatomical features common to all tumors. No parasitic origin can be suggested for any of these excepting carcinoma; and in view of this fact, and in view of the fact that the evidences of the parasitic etiology of carcinoma are so imperfect, we must still regard the new growths as either belonging to the class of misdevelopments, or as being due to some cause exciting the fixed cells to a greater than normal activity.

The most plausible of parasitic theories is that of the sporozoa. Its right to consideration as an etiological factor of carcinoma is, however, nullified by the fact that its champions are contradicted by an equal number of equally accurate

observers. The doctrine of the bacterial cause is to-day quite unchampioned; and, finally the theory of Adamkiewicz is unnatural and illogical. The hypothesis of Cohnheim, though not histologically substantiated, still stands as accounting for many inexplicable phenomena. It is, however, within the range of the greatest probability that the cause of tumors has been sought for too wide away from the real nature of things, and that it lies nearer the ordinary and constantly operating phenomena, and not among the rarer and more remote biological possibilities. In such simple growths as clavus no far-fetched etiology is sought, but simple mechanical irritation satisfactorily accounts for the piling up of epithelial cells. May we not presume, then, that chemical or mechanical irritation, per se, may be a cause of carcinomatous degeneration without complicating the etiology still more by the hypothesis that such irritation is a cause only as it impairs the cell vitality and renders it an easier prey for the specific microorganism? The expression "irritation," for want of a better term, must stand for inflammatory processes giving rise to parenchymatous or interstitial cellular changes.

Mr. Darwin has shown that habits or induced conditions acting continuously through a number of generations finally become engrafted upon the line, and are transmitted as instincts or inherent qualities; and it is not possible that epithelial cells, which for a series of generations have propagated under such abnormal conditions as have caused and as exist in a primary carcinoma, may become endowed with the pernicious vitality which finally characterizes the metastatic growth?-Annals of Surgery.

OBSTETRICS.

CESAREAN SECTION IN PLACENTA PREVIA.

DR. FORD formulates the following conclusions: (1) The dangers of placenta prævia, as well to the mother as to the child, are due to the development of the placenta upon the lower uterine segment, and to the canalization of this segment during labor. (2) While the first of these conditions cannot be avoided, the second should not be permitted in placenta prævia totalis or partialis. Delivery should be by Cæsarean section. (3) In placenta prævia marginalis, if the circumstances were favorable, the os easily dilatable, the condition of the mother and child good, the head presenting or capable of being steadily brought to engage, and the hæmorrhage arrested or moderate, it would be well to follow the method of intrauterine and vaginal tamponade, and deliver by forceps if the child should be in danger. But if the os were rigid, the hemorrhage profuse, the presentation lateral, the cord prolapsed and not reducible, or the foetus evidently suffering, immediate recourse to Cæsarean section should be had. (4) The Cæsarean section should be performed as soon as the diagnosis is established and the condition of the mother permits, to the exclusion of all other methods, as an elective and primary operation, in all cases of placenta prævia totalis and partialis, and as soon as the conditions warranting it, in placenta prævia marginalis, have been satisfactorily determined. (5) In the two graver forms of placenta prævia the Cæsarean section should be practiced as a prophylactic measure, in place of any attempt to deliver by the natural passages, after the first hæmor

rhage. (6) In cases where hæmorrhage is late or sets in only as labor begins, and where consequently, the placenta is most probably attached laterally, it is advisable, until this entire subject has been practically studied, to deliver per vaginam as a rule. If, therefore, the cervix be easily dilatable, and the hæmorrhage moderate, we may proceed as suggested in the more hopeful cases of marginal implantation. But even here, an undilated os associated with severe. hæmorrhages would constitute a very serious condition. If the rigidity were due to fibrosis, it should be abated by multiple incisions; if to carcinoma, the radical Cæsarean section would be indicated. If the cord were prolapsed, and after reposition still descended, the os being partly dilated and not dilatable, dangerous hæmorrhage continuing meanwhile, the Cæsarean section would be unquestionably indicated for the safety of both mother and child.-American Gynecological Journal.

PATHOLOGY.

INFLUENCE OF WATER RICH IN SALT UPON THE DEVELOPMENT OF CHOLERA BACILLI.

AUFRECHT (Centralbl f. Bakt., Band XIII, Numbers XI and XII) found that the refuse water of certain soda and potash works caused the percentage of salt in Elbe water to be considerably raised. Investigating the bearing of this fact upon the development of cholera, he found that comma bacilli grow as well upon nutrient gelatine made with this water as upon gelatine to which crystallized soda in the proportion of one per cent. has been added, as recommended by Dahmen. This observer has shown that cholera bacilli grow much better upon the latter medium than upon the ordinary alkaline gelatine. It appears, therefore, that the Elbe water strongly charged with salt favored the development of cholera-Trenkmann (ibid., Band XIII, number X, shows that the development of cholera bacilli in sterilized spring-water is much favored by the addition of salts of soda and potash to the water. In unsterilized spring-water the nature of the salt added has an important bearing upon the development of saprophytes and comma bacilli respectively. Thus, the addition of common salt favors the development of the former, whilst if a mixture of common salt and sulphate of soda be added a great increase of cholera bacilli is evident, in contrast with which is the comparatively slight increase of saprophytes; further, several varities of the latter disappear altogether.-British Medical Journal.

LEUCOCYTES IN INFECTION AND IMMUNIZATION.

MLLE. EVERARD AND MM. DEMOORE AND MASSART (Ann. de. 'l Inst. Pasteur, February, 1893) have made an experimental study of the quantitative and qualitative variations of the leucocytes during infection and immunization. They describe three forms of leucocytes. The first are small, with a single, compact nucleus which stains deeply with hæmatoxyline; the protoplasm is small in amount, and stains slightly with eosin. The second are large leucocytes, with a single vesicular nucleus which stains feebly with hematoxylin, and is spherical, reniform, horseshoe-shaped, or even annular in form; the protoplasm is abundant and homogeneous, staining freely with eosin. The third are large leucocytes,